shunsukeaihara/pydtw

Fast Imprementation of Dynamic Time Warping For Python

pydtw's Issues

Reference required

Can the author please specify a reference to the psuedoalgorithm or the theory behind the implementations, especially 2D version? Thank you

return value of cost?

Is there a way to get a single floating point value for cost or distance versus an ndarray?

impliment LB_Keogh

[SOLVED] setup failed on newer macos i.e. 10.14 (mojave)

if you are newer macos user, you may encounter error like this when using pip install,

pydtw/dtw.cpp:584:10: fatal error: 'ios' file not found
#include "ios"
         ^~~~~
2 warnings and 1 error generated.
error: command 'gcc' failed with exit status 1

the solution is to add two line in setup.py after imports

import os 
os.environ['MACOSX_DEPLOYMENT_TARGET'] = '10.9'

reference :
pandas-dev/pandas#23424 (comment)

ImportError: No module named 'Cython'

Pip install pydtw throws import error.

Collecting pydtw
  Downloading pydtw-1.0.0.tar.gz
    Complete output from command python setup.py egg_info:
    Traceback (most recent call last):
      File "<string>", line 1, in <module>
      File "/tmp/pip-build-16ffmgc7/pydtw/setup.py", line 3, in <module>
        from Cython.Distutils import build_ext
    ImportError: No module named 'Cython'
    
    ----------------------------------------
Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-build-16ffmgc7/pydtw/

Possible bug?

x1 = np.array([1,2.0,3])
x2 = np.array([1,1.2,3])
dist, cost, a, b = dtw.dtw1d(x1, x2, )
print(a)
print(b)

returns
[0, 0, 1, 2]
[0, 0, 1, 2]

Shouldn't this be:
[0, 1, 2]
[0, 1, 2]

Using pydtw 2.0.2, Python 3.8.1, numpy 1.18.1.